Flight Dynamics Recorder ?

something between worthless and ridiculous :)

just the data. preferably in an easy to read format, that could be massaged by 3rd party software. make the format self explanitory so some of us can cobble up our own software.

all :) plus a couple of screw terminals for us to work up our own sensors.

I'm not sure alot of people would be interested in x and y accelleration data. Especially if the rocket is spinning, it's pretty much useless (and will be very small by comparison to the z-axis). A single axis sensor would be cheaper too.

I'm not sure temperature is that important, unless it's going to be a really high flight that exceeds a minute (it will take the sensor a while to measure the transient response). Now a temperature sensor on the nose cone (to measure aerodynamic heating) would be alot nicer for the really fast rockets.

For acceleration, 50 samples/sec MINIMUM. 100 is better. For pressure, 10/sec is probably adequate.

I think excel covers the graphing capability.

Spin rate would be nice. And the other things that could tell you which way the rocket was pointed (sun sensor, earth/horizon sensor, magnetometer, etc).

I'm working on a datalogger now that covers acceleration, pressure, and a horizon sensor (which is the hard one to make). Total cost so far has been about 70$, although I got the accelerometer as a free sample. It's got an eeprom dip socket, so you can get up to 256kb of data (but I'm sure 8kb will be adequate).

Dave

Surface measurement?

I dont understand this. If its traveling in a verticle direction, where will you launch it that there will be no gravity?

Will this be installed on the center of rotation of the airframe? Or will this be off center? What effects on the data collection will there be with an off axis sensor? I was hoping to answer these questions myself, but my 3 axis acceleration sensor did not collect any data this past weekend. Windows was running and it told me it started recording, but no data. I even bench tested it before and after the flight, recorded data just fine. Hmmmm.

The biggest cost is memory, you need a place to stuff all this data. Think magnetic tape ;-)

I assume it will have a UART port for data download, just simply spew the data to this port durring flight, then the user can cobble up his own TX/RX and have telemetry.

If you are recording many channels at a high sample rate, then people will want it to download as fast as possible. Just download as fast as the UART can handle.

Graphing software that will export to CSV is the way to go. But in a pinch you can use Hyperterm, works great, and no software to write. Just have your unit download the data in CSV format.

I dont know about Electric fields. Sound? Umm thats either going to require lots of space, or a microcassette recorder.

I was trying to design an X-ray detector. Wanted to use background X-ray radiation as an apogee detection mechanizm. I might still do it, but even the SQUID sensor requires LN2. Maybe I can get my hands on one of them X-ray lenses that were developed for the Starwars program. I have seen them used to collect/direct x-rays in a wavelength spectrometer. Hmmm Still need LN2 though. But I could use the boil off from the dewar to cool the power supply for the computer ;-)

RDH8

Ya, so with Hyperterm to download and Excel to display, your all set.

Dave, I assume you are talking about the 24xx256 memory that stores 256K bit of data. Microchip has the new 24xx515 that stores 512K bit now. Just watch out for that pin 3, it works a bit different than the 256 chip. Other than that its a drop in replacement. Actually being two banks, I assume its just two 256 die's in one package. I am working on some code to push it to store data at 2kHz. ;-)

RDH8

I was looking at a similar application with the Atmel nano-processors (ok, their microcontrollers, but I like the word nano-processor better). They have chips that go down to the 8 pin surface mount size. Won't hold a lot of data (64 bytes), but you could put one in a bt-5! Great for only recording max altitude & time, etc.

One idea I was thinking about was doing Run Length Encoding on the data. Therefore, the limits become, not time, but delta (change) of the data. Theoretically, if you have the thing running on the pad for

30 minutes before launch, it would only chew up 1 slot worth of memory.

Robert,

DId you ever do a version of your controller board with the AVR?

I have a couple timer applications (airstarts) and it would be a lot easier if I could use off the shelf boards. Your controller and high-drive cards are just the ticket, but I'm a 6800 snob who can't bring himself to do PIC. OTOH, I do have some ancient AVR tools at the house. (As well as some HC05 tools :)

Doug

Doug, I did make an AVR based Roctronics board. But only one.

If you would like one let me know and we can work something out.

I do have a tiny little timer at

;-)

RDH8

I use the 8pin PIC variant of that. Works great, and tiny. I put one in my Estes Mosquito ;-) You can see that here:

's in the gallery.

RDH8

Sweet. What do you use for a sensor?

I am using a SX15A pressure sensor, but I'm unhappy with the need for a separate amplifier. I was hoping the 20x amp built into the Atmel would be enough, but the bias on the SX15A consumes quite a bit of the ADC bit space.

Look at the Motorola sensors. The MPX4115 is internally calibrated and compensated. So no external amp required, output goes right into the A/D.

Yeah, nice, cheap, and super-easy to integrate. I have the 6115. Goes to 2.7psi. Very similar to the 4XX5's though.

Dave

How is it different and better than this?

I use these in student rocketry projects, it is excellent. Sampling rate is adjustable, I use 200 samples/second.

John Kallend, Professor, Mechanical, Materials and Aerospace Eng> Greetings Friends,